Improvements in Polarographic Instrumentation G. E. PHILBROOK' AND H. AI. GRUBB Standard Oil Co. (Indiana), Whiting,Znd.
The Sargent Model XI1 polarograph in this laboratory contains a Rubicon taut suspension galvanometer with a sensitivity of 0.003 microampere per mm. at 61-cm. (2-foot) scale distance. This galvanometer is very sensitive to vibration. A vibration-isolating mount constructed for the instrument successfully remedied a vibration problem which according to the makers is the worst they have seen. To make full use of compensation technique some means of damping the excessive galvanometer oscillations at high sensitivity is required. A n electrical filter avoids many of the drawbacks of the circuits previously described for this purpose. A polarograph cell has been designed which appears more satisfactory for many routine uses than those described in the literature.
E
SCESSIVELY severe vibration from railway traffic prevented the operation of a Sargent-Heyrovskj. Model XI1 polarograph equipped with a Rubicon galvanometer. When the instrument stands on a substantial laboratory bench the galvanometer vibrates continuously with an amplitude of 0.5 to 2 mm. Occasional shocks cause deflections of 10 to 15 mm. Mounting the instrument on rubber, slate and rubber, or floating it on mercury increased the instability. A short period pendulum mount appeared to give a very slight improvement. Placing the instrument on a pad of 5 cm. (2 inches) of felt covered with 31.75 kg. (70 pounds) of sheet lead gave sufficient improvement to make 50% of the records usable.
In Figure 1 are shown two vibration records, which were obtained by substituting a resistor for the cell and scanning the voltage range. The first record was made with the instrument resting on a substantial and rigid laboratory bench. The second curve was made under comparable vibration conditions with the instrument in its mounting. I n Figure 2 is shown a schematic cross section of the mount. The instrument rests on a 0.75-inch plywood bed plate, supported on two U-shaped yokes spaced 11.5 inches on centers from front to back. The yokes are 8 inches deep with an over-all !ength of 22 inches and are fabricated from 0.25 X 2 inch strap iron. The ends of the yokes are turned down to give a horizontal section 2.75 inches long, which is drilled to take the mounts. B 4-pound Lord plate-form rubber-in-shear mount (Lord Manufacturing Co., Erie, Pa., Catalog No. 102PH4) is bolted to the bottom of each of the horizontal yoke sections. The rubber mounts are on 20-inch centers from side to side and 11.5inch centers from front to back. The upper rubber mounts of the support pairs are the same as the lower and are bolted to straps of 0.25 x 2 inch flat iron. The yokes are suspended from these upper mounts by 2.5-inch KO, 8 machine bolts through the center holes. All straps and yokes are pierced with 1-inch holes opposite the mount centers t o clear the bolts and rubber which stretches greatly under the high overload imposed. The straps are supported on a heavy wooden frame fabricated from 2-inch pine with mortised joints. Sections are cut out a t the suspension points t o clear the suspension members. The front is open and a section is cut out a t the right end to clear the camera. I n Figure 3 is shown a vertical view of the mount with the instrument and bed plate removed. The dash pot is 4 inches off center at an angle of 45' to the center lines of the mounts. This location was chosen to damp rotation as well as lateral motion.
RESTS ON FRAME
Figure 1. Vibration Records 1. Instrument on laboratory bench 11. Instrument in mount
+
cG
OF INSTRUMENT
The system now in use employs four pairs of commercial rubber-in-shear mounts overloaded 300%. The instrument is suspended from these units with its center of gravity close to the plane of the centers of the suspension pairs. This technique reduces pendulum effects to a minimum. A dash pot mounted away from the center of the bed plate, and filled with a very viscous oil, rapidly damps oscillations of the suspension started by manipulation of the instrument. A cam-operated lift plate takes part of the load off the suspension when the instrument is not in use. 1
Ga.
Present address, Chemistry Department, University of Georgia, Athens,
Figure 2.
Schematic Cross Section of Polarograph Mount
...............
- I ~ ~ ~ .
................
. . . . . . . .
A N A L Y T I C A L CHEMISTRY
8
A 2inoh can open a t the bottom and fastened to the bed plate dips into the dash pot, which is filled with very heavy oil (Penn sylvania steam-refined, viscosity 220 scconds Saybolt a t 210' F or equivalent). The can is immersed 2 inches in the oil. Thi cam-operated lift d a t e shown in tho center has 5 vertical rise o 0.25 inch. c,,cy,oa, Mra
GALVANOMETER OSCILLATION DAMPER
Linganc and Kerlinger (3) point out that compensation technique is limited by wide oscillations of the galvanometer when the current being compensated reaches a value 10 t o 15 times larger than the current being determined, and suggest the use of an electrolytic condenser of fairly high capacity acro8s the Ayrton shunt output to damp these oscillations. To cover a wide range of concentrations several condensers are required, as there is an optimum capacity for each combination of sensitivity and drop time. To overcome this drawhack Fill and Stock (1) devised an improved circuit in which a condenser was used across the Ayrton shunt output and a variable resistance in series u-ith the Ayrton shunt center tap. This circuit damps effectively over a wide range of sensitivity-drop time values. Fipure 4. Wiring Diagram for Modified Sargent-Heyrovsk? Folarograph Model XI1
design considerationswere worked out and are presented h e l m For those who may urish to add such a circuit to other polarographs.
_ _ .......................................... _ .L" _ll/J ' . . I : ......... . . . . UI l l r . :.. ... ulleob ( i U l l t i l l b ILicIbv1LIIce, TCC,bb&llCt:
. _ I
.,,,IC,. . .
ill", U' lillc- 'Lb
.....
ulbrr..
is equal t o the sum OF the resistance of the galvanometer with its arspcisted Ayrton shunt and the 2200-ohm resistor placed in series. (The actual resistance of most radio-type resistors may de,.iate 20% from the mted d u e . This resistor was rated a t 2000 ohms.) The resistance of the galvanometer-Ayrton shunt combination varies from ahout 1ohm to a maximum OF about 300 ohms. The oapacity arm of the filter consists of two 1000-mfd. &volt electrolytio condensers in series opposing. The effcctive capacity of two identical condensers used in series in this manner is hnlf that of B single condenser. However, the capacity of a n electrolytio condenser is strongly dependent on the applied voltage. Measurement by ballisticmethodsgavoavalueof 1250-mfd effective capscity For the pair connected in series a t a voltage of 64 mv. This voltage corresponds roughly to normal conditions of u6e in the dzmper. If it is assumed that the reaotance of thc Figure 3. Vertical View oC Mount Instrument and bed plate removed
D=2- t In both circuits the use of ordinary electrolytic condensers leads to serious ditliculty. The majority of electrolytic condensers oommercially available contain internal e.m.f.'s of the order of 6 to 60 mv. Such potentials are sufioient to cause full-scale deflections of thc galvanometer. While high-capacity paper eorrdensers suitable for. such damping circuits have been manufactured commercially, they are too bulky to he installed in the Sargent-Heyrovskj. Model XI1 polarograph. The cirouit described is similar to that of Fill and Stock but avoids trouble due tointernsl e.m.f.'s of electrolytic condensers. The space requirement is approximately 1 X 2 X 2 inches. A resistor (R,, Figure 4) is placed in series with the low resistsnec end of the Ayrton shunt and the anode. Two electrolytic condensers (C, and C,, Fiwre 4) connected back to back are placed
cally identfcal with the arrangement of Fill and Stock, Since eleotrolytic condensers conduct current in only one direction, t.his arrangement Mocks the circuit to continuous direct current flow and thc e.m.f. of the condensers is inoperative. The oscillation damper consists of a simple resistance-capacity filter. The proper' cirouit values for the authors' instrument were obtained by trial. The damper performed so a e l l that the
7
similar to (h&
RC
shown in F h e 5. With the Avrton shunt set
dlmping factor 'is 3.8
D = 2a x 2220 x 1250 x
10-5 = 022
while the observed damping Factor is 0.18. At first glance it may appear that smaller damping factors are desirable. While there is considerable choice in the damping factor employed, this technique has definite limitations. Damping is achieved over a wide range of conditions a t the espense of introducing an iR drop in series with the oell. This alters the Form of the waves and a correct.ian For this effect must be made in the dctermination of wave slopes a d half-wave potentials. For small damping factors B second effect can he troublesome. The offect OF the damper is to increase the efrective period of the gdvlvanometer. If this increase in period is large enough, the galvanometer will lag so Far hehind the potential changes that
9
V O L U M E 19. N O . 1, J A N U A R Y 1 9 4 7
The one on the voltage control was removed and the second was moved from the cmrent control into series with the battery lead. I n Figure 4 is shown the circuit diagram for the instrument after all thc abqve modificationsu-ere madc. I n Figure 5 are shown four curves obtained on a solution 0.005
M in cadmium, and 0.0005 M in zinc in 0.1 N potassium chloride
Figure 5. Curves Obtained on Solution 0.005 M in Cadmium and 0.0005 M in Zinc in 0.1 N Potassium Chloride Containing Trace of Hydrochloric -4eid r
0.001%mrthyl red 4-".iri"ir"
containing a trace of hydrochloric acid. Methyl red (0.001'%) was used as maximum suppressor. Curve 1 was obtained a t a sensitivity of 1.00without the damper. Curve 2 was obtained a t the same sensitivity with the damper in operation. Curve 3 shows the zinc wave at a sensitivity of 20 after compensation without the damper. Curve 4 was obtained with the damper under the same oonditions as curve 3. The small irregularities in curves I, 111, and IV of Figure 5 are believed to be due to vibration of the capillary. The cell and capillary are attached to a rigid support which rests on a felt pad. The support is not connected to the wall. A mount for the cell assembly based on the principles used for the instrument does not appear to he practical, because of constructional difficulties. The small irregularities shown do not appear appreciably to affect the results obtained with the instrument. POLAROGRAPH CELL FOR ROUTINE USE
I"" .."rl.-"-rl
waves will show little or no straight portion on the rise or plateau. This trouble can be reduced partially by decreasing the rate of inereasemf potential to the lowest value which gives satisfactory breaks. The size of damping factor should be selected with the possible uses of the instrument in mind. This damper has been in u ~ on e the authors' instrument for over nine months and has proved extremely satisfactory. T+ in=*rlln,4 -i*h 9 double poledouble thro from the circuit if desirs on all wave3 because t he tions on normal waves gb precision of measurem en In order to realize tb e
The polarograph cell for routine use by nontechnical personnel should be constructed from standard items replacea.hl.ble from stock in case of breakage. As many components as possible should be fixed in place t o reduce manual manipulation t o a. minimum. Accessories not in routine use, such as reference electrodes, should be omitted. I n Figure 6 is shown a cell which the authors believe meets these requirements far most routine u8es better than any they have found described in the literature, The construction is evi-
n
n
As originally supplied the compensator voltage control WBB a 1000-ohm radio-tvue Dotentiometer with a 270' rotatition. ThiLunlt (Q) mas replaced with a 5000-ohm l0-turn (3600" rotation) Helipot (Helipot Corp., 1011 Mission St., South Pasadena, Calif., Catdog No. 680-H). This potentiometer extends the range of the compensator 37% and, more important, avoids the jumpy behavior apparent in the upper 20% of the old oomuensator voltaee ranee. The original compemhor had two &ut-offUswitches.
SCHEMATIC ELEVATION NO 3 C I L E
Figure 6. Polarograph Cell
10
I
dent from the drawing and requires no comment. A standpipe of the type described by Lingane and Laitinen ( 4 )is used with this cell. If desired, a silversilver chloride electrode of the type described by Lingane ( 2 ) may be used in place of the mefcury pool. This cell has been used for acid, alkaline, and neutral media for several months and no trace of contamination from the stainless steel capillaries has been detected. This convenient, compact, and rugged cell has proved itself in continuous service for several months on one capillary.
ANALYTICAL CHEMISTRY ACKNOWLEDGMENT
The authors would like to express their appreciation to T. H. Rogers, E. )I7. Thiele, and E. B. Tucker for their advice and interest. LITERATURE CITED
(1) Fill and Stock, Trans. Faraday Soc., 40,502 (1944). ANAL.ED.,16, 329 (1944). (2) Ljngane, IND. ENQ.CHEM., (3) Lingane and Kerlinger, Ibid., 16, 750 (1944). (4) Lingane and Laitinen, Ibid., 11,504 (1939).
Dielectric Identity Test for Plasticizers Polyvinyl Chloride Plastics Type MYRONA. ELLIOTT, A. RUSSELLJONES, AND LUTHER B. LOCKHART Naval Research Laboratory, Washington,D. C.
A new type of cell and method are described for the rapid determination of the dielectric properties of small samples of plasticizers as a function of temperature. Since these materials have characteristic loss factor-temperature curves that are sensitive to contamination by significant amounts of practically any foreign material, such dielectric data are valuable as an identity or acceptance test. Although the procedure has been applied to some plasticizers in particular, it may be used on nearly all electrically
B
ECAUSE of shortages of materials during the past few years,
it has sometimes been difficult to maintain uniform quality in manufactured products. One likely cause for variation in the electrical and mechanical properties of plastics made from the vinyl chloride polymers was variation in the plasticizers. The electrical method, as described herein, was found to be a convenient and reliable procedure for establishing the practical identity of a plasticizer from a given polyvinyl chloride product with that from previously manufactured batches of the same product. I t is a comparison method whereby the unknown is compared to a standard. The method is not sensitive to traces of impurities but will indicate the presence of foreign materials in a plasticizer if they are in quantities sufficiently great to affect the mechanical or electrical properties of the finished plastic. This method is not proposed as a general analytical tool, but it is recommended for identity and acceptance test purposes. It should also be useful as a control test in plasticiaer manufacture and in the manufacture of chlorinated or other polar organic liquids. No general scheme has yet been devised for the systematic analysis of plasticizers. The conventional chemical tests are used for phosphorus, chlorine, phthalate, etc. Physical properties such as refractive index and specific gravity are also an aid in identifying these materials, In this work, advantage is taken of the above conventional procedures in devising methods of identification for the kind of plasticizers found in the electrical insulation of synthetic resin-insulated Kavy cables. Advantage is also taken of a new method which is based on the fact that each plasticizer or mixture of plasticizers shows a characteristic variation of dielectric constant and loss factor (anomalous dispersion) with change in temperature when the measurements are made at high radio frequencies. This method is new only in its application as 1 Present
address, University of Wisconsin, Madison, Wis.
polar organic liquids. The method involves the extraction of a small quantity of the plasticizer from a resin and then the determination of the electrical properties of this plasticizer at a frequency of 10 &IC. as a function of temperature. Each type of plasticizer gave a reproducible curve different from that of the others. Simple qualitative chemical tests were found valuable for confirmation purposes on some types of plasticizers. Refractive index determinations gave useful supplemental information.
an identity test for plasticizers and other organic liquids. The anomalous dispersion of the dielectric constant is a well-known phenomenon and is discussed in many publications (1-4, 6, 9). Because of the polar nature of plasticizers, electrical methods are well adapted t o their characterization. The only problem here is the development of methods that are simple and rapid enough to be used in a routine testing laboratory. THEORY
All the plasticizers used with polyvinyl chloride fall into a general class of liquids that are polar but not ionized. When such liquids are placed in an electrostatic field, the molecules tend to become oriented because of electrical interaction between the electric dipoles in the molecules and the field. When this field is alternating, the molecules tend to follow it and so are turned first in one direction and then in the other. If the alternations of the field are sufficiently rapid, the molecules can no longer follow, but remain in a random state. Under these high-frequency conditions the liquid has the low dielectric constant and low loss factor of a nonpolar medium. At intermediate frequencies where the mole cules can become partly oriented, the dielectric constant has an intermediate value while the dielectric loss in the liquid is a t a maximum or peak. At low frequencies where the molecules become oriented to the maximum extent their thermal motion permits, the dielectric constant reaches its highest value (static value), while the loss factor drops to a low value again. This anomalous dispersion of the dielectric constant was first observed by Drude in 1897. In 1913 Debye gave the theoretical interpretation of this phenomenon that is generally accepted today. Since the application of the Debye theory to the anomalous dispersion of the dielectric constant of liquids has been given in detail (I,2 ) , it is not necessary to give an outline here.
